UNITED STATES ENVIRONMENTAL PROTECTION AGENCY

                          WASHINGTON, D.C. 20460
July 20, 1988

The Honorable Lee M. Thomas                          .    THe ^"'
Administrator
U.S. Environmental Protection Agency
401 M. Street, s.W.
Washington, O.C.  20460

Dear Mr. Thomas:

     The Water Quality Advisories Subcommittee of the Science
Advisory Board's Environmental Effects, Transport and Fate
Committee has completed its review of the office of Water's
guidelines for preparing water quality advisories.  !The review
was requested by Edmund M. Notzon, Director of the Criteria and
Standards Division of the Office of Water and the review was
conducted on October 22 and 23, in Annapolis, Maryland.

     The Subcommittee recognizes the potential that, the advisory
concept provides for bringing a greater number of pollutants
under regulatory control in a relatively short period of time.
The concept represents a preliminary step towards water quality
criteria development and is designed to protect both ambient
aquatic life and human health.  The primary issue regarding
ambient aquatic life protection involves defining and obtaining a
minimum data base for advisory development. Data describing
interactions in aquatic systems have not been developed for many
pollutants.  Data are more prevalent for characterizing human
health risks, and obtaining a minimum data base is not of
concern.  Instead, the primary issue for human health protection vxa
advisories is the appropriate depth of review of the existing
data base.

       In general, the Subcommittee has more support for the
concept as it applies to ambient aquatic life protection than
for application to human health protection, based on availability
of data.  Subcommittee cautions that advisories should not
substitute for water quality criteria and recommends that a
mechanism for advancing advisories to criteria status be
specified to insure that emerging data are enfolded to reduce
uncertainty,  in addition, the Subcommittee feels that it is
imperative that a review process and public comment period be
incorporated to balance the increased uncertainty inherent in the
advisory process.

     Regarding advisories for the protection of ambient aquatic
life, an approach which the Subcommittee endorses, some
suggestions for improving the guideline documents and the process

-------
                                                       Page 2

in general ar* provided.  These suggestions includes  specifying
the'data ftMttod, to advance understanding of the toxicity of the
pollutant aHressed, documenting uncertainty factors, employing
quality ratings, and including modifications to address exposure
duration, site-specific issues, and sensitive species rather than
those that are commercially or recreationally important.

     To guide decisions about human health, the Subcommittee
prefers the use of the water quality criteria process, and has
reservations with applying the advisory concept.  Particular
opposition is directed toward the restriction of literature
searching to the most recent 5 years, and to the use of secondary
literature sources.  The risk assessment procedure described in
the guideline documents focuses on data presentation rather than
on analysis, and the use of modifying factors cannot be endorsed.
Most of the Subcommittee's criticism is related to the- error that
results in assuming that a short risk assessment is easier,
quicker and less expensive than a long one.  In fact a properly
prepared, concise advisory could take less paper, but require the
same level of effort required for a criteria document.

     The Subcommittee appreciates the opportunity to conduct this
scientific review.  We request that the Agency formally respond
to the scientific advice transmitted in the attached report.
                                    Sincerely
                                                I
                                    Norton Nelson, Chairman
                                    Executive Committee
                                    Science Adyisorv Board
                                           rtung,
                                    Environmental Effects,
                                      Transport and Fate
                                                 -U
                                    Kenneth Dickson, Chairman
                                    Water Quality Advisories
                                      Subcommittee
Enclosure
cc:  A. James Barnes
     Lawrence Jensen
     J, Michael Conlon
     Edmund Notzon
     Frank Gostomski
     Donald Barnes

-------
    United States      Office of the Administrator ' 3A5-EET&FO-88-032
    Environmental Protection  Science Advisory Board  ^ly, 1988
    Agency         Washington, D.C. 20460 Final Report
'*•   Report of the Environmental
    Effects,  Transport and Fate
    Committee
       Review of the Guidelines
       for Preparing
       Water Quality Advisories

-------
              U.S. ENYIRONMSNTAL PROTECTION AGENCY

                             NOTICE


     This report has been written as a part of the activities of
the Science Advisory Board, a public advisory group providing
extramural scientific information and advice to the Administrator
and other officials of the Environmental Protection Agency.  The
Board is structured to provide a balanced expert assessment of
scientific matters related to problems facing the Agency.  This
report has not been reviewed for approval by the Agency; and,
hence, the contents of this report do not necessarily represent
the views and policies of the Environmental Protection Agency or
other agencies in Federal government.  Mention of trade names or
commercial products does not constitute a recommendation for use.

-------
              U.S.  ENVIRONMENTAL PROTECTION AGENCY
                     SCIENCE ADVISORY BOARD
       ENVIRONMENTAL EFFECTS, TRANSPORT AND FATE COMMITTEE
              WATER QUALITY ADVISORIES SUBCOMMITTEE

                             ROSTER

chair

     Dr. Kenneth Dickson
          Institute of Applied Sciences
          North Texas State University
          P.O. BOX 13078
          Denton, Texas  76203

Vice Chairman

     Dr. Rolf Hartung
          Professor of Environmental Toxicology-
          School of Public Health
          University of Michigan
          Ann Arbor, Michigan  48109

Members

     Dr. Ira Adelman
          Department of Fisheries and Wildlife
          University of Minnesota Twin Cities
          1980 Folwell Avenue
          St. Paul, Minnesota  55108

     Dr. Herb Cornish
          Professor Emeritus
          University of Michigan
          830 West Clark Road
          Ypsilanti, Michigan  48198

     Dr. Patrick Durkin
          Syracuse Research Corporation
          Merrill Lane
          Syracuse, New York  13210-4080

     Dr. Ron Jarman
          ERT
          12655 North Central Expressway
          Suite 706
          Dallas, Texas  75243

     Dr. Richard Kimerle
          Monsanto Corporation
          800 N. Lindbergh Boulevard
          St. Louis, Missouri  63167-5842

                               ii

-------
     Dr»  Dan Menzel
          Director and Professor
          Pharmacology and Medicine
          Director
          Cancer Toxicology and Carcinogenesis Program
          103 Jones Building/Box 3813
          Duke University Medical Center
          Durham, North Carolina  27710

     Dr.  BrocK Neely
          InviroSoft
          4302 Cruz Drive
          Midland, Michigan  48640

     Dr.  John Neuhold
          Department of Fisheries and Wildlife
          College of Natural Resources
          Utah State University
          Logan, Utah  84322

     Dr.  Verne Ray
          Medical Research Laboratory
          Pfitzer, Inc.
          Groton, Connecticut  06340

     Dr*  Thomas Waller
          Program in Environmental Sciences
          University of Texas at Dallas
          P.O. Box 688
          M.S. BE-22
          Richardson, Texas  75080
Support
     Ms. Janis C, Kurtz
          Executive Secretary
          Science Advisory Board
          U,S, Environmental Protection Agency
          401 M Street, S.W.
          Washington, D,C,  20460

     Mrs, Lutithia V. Barbee
          Secretary to the Executive Secretary
                               ill

-------
1.0  EXECUTIVE SUMMARY

     Public pressure for control of pollutants,  and the lack of
resources to support the water quality criteria-setting process
traditionally used for pollutant control have led EPA to propose
the water quality advisory concept for protection of both ambient
aquatic and human health.  A water quality advisory (WQA) is a
numeric recommendation that provides an estimate of the pollutant
concentration in water which is unlikely to result in adverse
effects to human health or the environment, even when exposures
are continued for a lifetime. A WQA is similar to a water quality
criteria (WQC) in that each is based on data describing
toxicological effects, however, an advisory is provided when a
lack of data prevents development of a WQC.  Application of this
concept clearly has the potential to bring a greater number of
pollutants under regulatory control in a relatively short time.
However, since advisories are based on fewer data, they are
accompanied by increased uncertainty*  The Subcommittee arrived
at recommendations and conclusions regarding the advisory
concept in general,  In addition, specific findings that address
the guidelines for ambient aquatic life and human health,
respectively, are provided.  The recommendations, conclusions and
findings are summarized below.

1.1  The Advisory Concept

     The Subcommittee endorses the advisory concept as it applies
to development of ambient aquatic life advisories, provided they
do not substitute for the development of water quality criteria.
The uncertainty attending the advisory concept may cause under or
over protection.  Therefore, the advisory process should include
mechanisms for reducing uncertainty as more data become
available.
     EPA should consider state standard-setting and management
practices, carefully evaluating the problems that may result if

-------
advisories are adopted as standards.   In states where the law
stipulates that water quality standards cannot be made less
stringent once they are adopted, problems may arise as advisories
progress to criteria that are based on more knowledge.

     A procedure for selection of chemicals for evaluation should
be incorporated into the advisory process.  The Subcommittee
recommends that the chemicals selected for advisory development
be those for which there is evidence of potentially significant
exposure in the environment.

     The Subcommittee feels it is imperative that a review
process and public comment period be incorporated into the
advisory process.  The increased certainty provided by the peer
review process is necessary to balance the increased uncertainty
inherent in the advisory process.

     In general, the Subcommittee had more support for the
concept of advisories applied to protection of ambient aquatic
life than for application to human health protection, primarily
based on the availability of data.  A considerable body of
evidence and expertise support procedures assessing risk to
humans.  The Subcommittee has reservations about basing
advisories on only a subset of these resources,  in contrast,
less data are currently available to support decisions required
to protect aquatic life.  Therefore, a program designed to
generate more data to support such assessments and fill in gaps
in the knowledge base can be readily endorsed by the
Subcommittee.  Specific findings follow.

1.2  Ambient Aquatic Life Advisories

     The Subcommittee endorses the concept of ambient aquatic
life advisories and suggests some modifications to motivate
advances in the area of aquatic effects prediction.  First, the
Subcommittee recommends that the Agency provide better guidance
in the supporting documents to ensure that any additional data

-------
collected In response to the advisory process also advances the
understanding of the toxicity of the chemical in question.

     In addition, the Agency needs to develop documentation to
demonstrate the basis of each uncertainty factor used in
developing an Advisory.  A procedure should be developed to rate
advisories on the basis of the quantity and quality of the data
used to calculate and advisory.  The value derived from the
rating system should be presented in the Advisory statement.

     The Agency should modify the guidelines to incorporate the
concept of exposure duration and the flexibility to account for
site-specific differences.  Another recommended modification is
determination based on effects produced in species that may be
the most sensitive rather than on species that may have
commercial or recreational importance.

     The Subcommittee endorses the development of Advisory
guidelines for specific chemical groups based on a justifiable
rationale, such as structural activity relationships.  Also, the
Agency is encouraged to develop Water Quality Criteria and/or
Advisories for wildlife protection.

1.3  Human Health Advisories

     As previously stated, the Subcommittee has reservations with
applying the advisory approach to human health protection.  The
Subcommittee has several recommendations that may improve the
advisory process.  For example, secondary sources, those that
summarize the primary journal literature, are recommended for use
only as directories to the primary literature, rather than as
data sources themselves.  Secondary synopses often omit pertinent
data in the course of summarization.

     The Subcommittee particularly opposes restricting the
supporting literature search for advisories to the most recent 5
years.  This restriction discourages the use of the complete body

-------
of information that is available, including older studies at the
foundation of toxicology, and studies that may report important
clinical findings or industrial exposures in humans»   In the
opinion of the Subcommittee, a responsible rislc assessment cannot
be conducted without thorough collection, review and
interpretation of all pertinent data.

     Guidelines for preparing human health advisories differ from
other risk assessment procedures in the use of modifying factors.
The necessity for "these factors is unclear, and the procedures
for their use are not well developed,* therefore, the Subcommittee
cannot endorse their application to the advisory process.  In
addition, the risk assessment procedure described in the
guidelines for human health advisories seems to emphasize
presentation of facts, rather than analyzing and comparing them.
Most of the Subcommittee's criticism is related to the error that
results in assuming that a short risk assessment is easier,
quicker and less expensive than a long one.  In fact a properly
prepared, concise advisory could take less paper, but require the
same level of effort required for a criteria document.

-------
2.0  INTRODUCTION

2.1  Origin of the Review

     On June 30, 1987, EPA's Office of Water requested that the
Science Advisory Board (SAB) review the draft guidelines
developed for preparing water quality advisories for both human
health and aquatic life protection.  SAB reviews are conducted
under the auspices of its Executive Committee.  On July 21, 1987,
the SAB received a preliminary briefing on the Water Quality
Advisory Guidelines, given by FranJc Gostomskl, Chief, Water
Quality Criteria Section, Office of Water Regulations and
Standards.  The SAB Executive Committee agreed to conduct the
review and delegated responsibility to the Environmental Effects,
Transport and Fate Committee, which established the Water Quality
Advisories Subcommittee to conduct the review and appointed Dr.
Kenneth L. Dickson as the chairman of the subcommittee.  The
request for this SAB review is presented in Appendix A.
2.2  Purpose of jbhe Review

     The purpose of the review is to provide an independent, peer
assessment of the scientific adequacy of the objectives, rational
and methodology included in the water quality advisory guideline
documents presented to the SAB.  The Subcommittee's objectives
are to evaluated the concepts underlying the water quality
advisory approach, and to provide the Agency with recommendations
and suggestions for improvement.

2.3  Review Procedure

     The Subcommittee received two guideline documents in advance
of their meeting! 1) Guidelines for the Preparation of Office of
Water Health Advisories, and 2) Guidelines for Deriving Ambient
Aquatic Life Advisory Concentrations.  These documents are
included as Appendix B and C, respectively.  In addition, the

-------
Subcommittee received some examples of applications of the
guidelines to specific chemicals,  including ambient aquatic water
quality advisories for xylene, styrene and tetrachloroethylene;
and human health effects advisories for hexachlorobenzene and
2,4,S-trichlorophenol.

     The Subcommittee met in public session on October 22 and 23,
1987 in the Laboratory Conference Room of EPA's Region 3f
Annapolis office at 839 Bestgate Road, Annapolis, Maryland.
David Sabock from EPA, Office of Water Regulations and Standards
described the Agency's need for review and intended application
of the Advisory guidelines.  More detailed information was
provided by Frank Gostomski.  Specific methodologies and
rationale for development of ambient aquatic advisory
concentrations were presented by Dave Hansen (ERL-Narragansett)
and "Tom Pureell (WQC Section) while health methodologies and case
studies were presented by Cindy Mullen (ECAO, Cincinnati).

     Following these briefings, the Subcommittee discussed the
principles underlying the derivation and use of advisory
concentrations in a generic manner.  This discussion was followed
by sessions addressing specific aquatic and health issues by
appropriate subgroups of the subcommittee.  These discussions
formed the basis for recommendations, suggestions and comments on
the advisory guidelines.

2.4  Description of This Report

     The Subcommittee report provides general conclusions with
regard to fundamental concepts, documenting their strengths and
weaknesses.  In addition, the report provides a discussion of
specific issues that were identified during the review process,
issues that pertain to both ambient aquatic advisories and health
effects advisories as well as the field of water quality itself
from National and state perspectives.

-------
3.0  THE ADVISORY CONCEPT

     Two forces — public pressure for control of the increasing
number of toxicants that cause concern, and laclc of funds to
produce the criteria documents that provide for control of these
toxicants — have pointed to the need for change in the water
quality criteria-setting process,  EP&'s solution has been to
develop the advisory concept toether with accompanying guidelines
for setting advisory concentrations.

     Ten years ago the Agency signed a consent decree that
obligated production of criteria documents for a list of some
sixty substances.  Even with court-granted continuances, criteria
have been developed for only about half of the substances on the
list.  Though the remainder are substances of concern, data are
not sufficient to support a full-fledged criteria effort and
substantial laboratory and field testing will be necessary to
obtain a complete data base.

     Since there are many toxicants for which inadequate data or
no data exist, application of the advisory concept would bring a
greatly increased number of toxicants under regulation in a
relatively short time.  However, these advisories would
necessarily be based on fewer data and, therefore, attended with
greater uncertainty.  The consequence of uncertainty is an
increase in the likelihood of over or, in some circumstances,
under protecting human health and/or ambient aquatic life.

     Such tradeoffs give rise to concern among members of the
Subcommittee.  These concerns translate to reservations with the
concept as expressed in the guidelines.  The Subcommittee offers
a series of conclusions and.recommendations regarding the
advisory concept in the.four sections to follow.

-------
3,1  Progressing from Advisories.. £_o_Cr_iter.i.a.

     There 'is no trigger within the advisory guidelines for
taking the concentration set by the advisory process to the
higher level of certainty provided by the criteria process, when
the etata do becoiae available*  Therefore, there is no motivation
to generate more data,, and no means to reduce "Uncertainty once an
advisory concentration is established.  The setting of advisory
concentrations is an appropriate first step in the criteria
setting process, however, it is not an appropriate final step for
either human health or arobient aquatic life protection.  The
Subcommittee believes that decision makers will be hampered with
tools that do not reflect the current state of scientific
knowledge and are, in addition, surrounded by uncertainty.

     The Subcommittee recommends that a well defined mechanism
for incorporating new data be included in the advisory
guidelines.  This mechanism should allow for an provide
incentives for elevation of advisory concentrations to more
certain and well founded criteria concentrations when sufficient
data exist.

3.2  Ant .ld.egr_edat, i on

     The WQAs are likely to be used in. a role roughly equivalent
to that of the WQCs in related management programs,  states are
likely to develop water quality standards and NPDES  (National
Pollution Discharge Elimination Systems) permit programs  around
water quality advisories, just as they currently do with  water
quality criteria.  However, there may be resistance to the
advisory concept in some states.

     Since EPA's guidance for developing advisories involves a
conservative approach, they result in a more stringent number.
If a WQC is subsequently developed, it is likely to be both more
certain and less stringent.  Some state Attorneys General have
interpreted state laws to mandate that no water quality standard
                           8

-------
may be made less stringent under the law.   Therefore,  even if EPA
replaces an advisory concentration with a  more certain but leas
stringent water quality criteria concentration, certain states
could be required to maintain overly stringent WQA as the water
quality standard.  For this reason, some states are likely to
resist adoption of a water quality standard based on an advisory,
and are likely to wait for EPA to issue subsequent criteria to
prevent establishment of overprotective standards.

3.3    Chemical selection

     Risk is a combination of exposure and effect.  If there is
no exposure to a chemical then even the most toxic chemical will
pose no risk.  Thus it would be a misuse of limited resources to
prepared a advisory document on a chemical for which there was no
established exposure.  The extent of exposure may be determined
through the use of on-going monitoring programs, such as those
used by the states in support of the permitting process.  For
these reasons, the Subcommittee recommends that chemicals be
selected for the advisory process only if there is evidence of a
significant exposure to the aquatic environment.

3.4  Peer Review and Public Comment

     The national guidelines for producing water quality criteria
documents mandate peer review and public comment.  The guidelines
for advisories do not.  The Subcommittee points out that this
process for producing advisories depends on the opinion of a few,
and believes that this would be a grave mistake.  In light of the
more restricted data base that will support advisories and the
increased uncertainty that a restricted data base provides, the
likelihood of producing an inappropriate advisory concentration
is greatly increased, however expert the opinion behind it.  The
Agency must outline a process that provides for review by a range
of external authorities in order to capture the best scientific
thinking, broaden the scientific consensus, and minimize future
criticism.  Therefore, the Subcommittee recommends that a

-------
procedure for peer review and public comment be incorporated into
the advisory process, and that such a procedure be described in
the guideline documents.
                           10

-------
4.0  AMBIENT AQUATIC LIFE ADVISORIES;  SPECIFIC ISSUES

4.1  Development, Certainty and Quality of the Data Base

4.1.1  Data Base Development

     The development of water quality criteria has required an
intensive review of the existing literature,  but has also been
characterized by the conduct of extensive laboratory research to
fill perceived data gaps.  From the proposed "Guidelines for
Deriving Ambient Aquatic Life Advisory Concentrations" (Appendix
B), it is clear that there will be dramatic differences in the
quantity of data available for establishing an advisory and that
the requirements for additional data will vary for each
substance-specific advisory.

     Given the potential for significant improvement in the
quality and quantity of data developed in support of advisories,
it is appropriate for the Agency to provide clear guidance to
ensure the efficient collection of such data.  For example, to
comply with current guidelines, a discharger may conduct yet
another generic acute toxicity test to strengthen the data base
for a chemical where three generic tests have been previously
performed, rather than producing data on species of particular
relevance that would be more informative or site specific.

     The guidelines recommend that dischargers seele guidance on
development of data from appropriate regulatory agencies.  The
Subcommittee recommends that dischargers be guided to develop
data that advance the understanding of the toxicity of the
chemical in question and that advisory documents stipulate the
advances that are needed to strengthen knowledge of the specific
chemicals that they address.
                           11

-------
4.1.2  Uncertainty

     Since the data base contains gaps, there is uncertainty
associated with every advisory estimated.   Ideally, the
uncertainty is quantified using a statistical,  probabilistic
approach.  In the example provided to the Subcommittee, a species
sensitivity factor of 11 and an acute to chronic ration of 25
were documented and used to estimate uncertainty.  The
Subcommittee recommends that additional documentation be provided
in guideline documents to clearly define these terms and state
their limits of applicability.  In addition, distinctions need to
be provided for application of these factors to fresh and salt
water.  Alternatively, generic uncertainty factors could be
tested to provide the necessary refinements in documentation.

4.1.3  Quality Indicators

     The Agency has recognized the differences in quality and
quantity of data that are likely to be encountered in preparing
advisories and has developed weighting factors based on the
nature of the available data.  Advisory documents will also
contain disclaimers, alerting users that they must consider the
technical basis of the advisory before application and that
advisory values are derived less stringently than criteria.  The
Subcommittee believes that a quality designation should also be
associated with the advisory concentration to indicate the
certainty or confidence attached to the advisory number, and the
fact that confidence varies.  Such quality designations could
take the form of descriptive statements or numerical indicators.

4.2  Considerations__for Modifying the Advisory Process

4.2.1  Exposure Duration

     As currently advanced a water quality advisory will consist
of a single concentration below which aquatic life are assumed to
be protected and above which adverse effects may occur.  The
                           12

-------
advisory concentration is designed to protect those exposed from
chronic effects and is analogous to the Criterion Continuous
Concentration (CCC) of a water quality criterion.  However, an
advisory does not contain the concept of exposure duration which
is an integral part of the CCC of a water quality criterion.  If
advisories are used as surrogates for criteria as stated in the
briefing to the Subcommittee, then the Agency should consider
modifying the advisory to include the concept of exposure
duration.  The CCC of a water quality criterion states that
ambient levels cannot exceed the CCC for more than four (4) days.
The inclusion of duration of exposure recognizes that effects are
a function of both concentration and time of exposure.  This well
established toxicological principal should be reflected in the
water quality advisories particularly if they are used for
regulatory purposes.

4.2.2  Species Beyond Commercially or Recreationally Important
         Ones

     The Subcommittee believes that a reduction in advisory
concentrations is warranted if laboratory or field data for any
species, regardless of whether or not it is considered
commercially or recreationally iaportant, indicate that it is
affected at concentrations below the calculated advisory
concentration.  Both the national water quality criteria and the
aquatic life advisory concentrations must be reduced if
"commercially or recreationally important species" are found to
be affected at lower levels than predicted by the final acute or
chronic values.  Restricting this adjustment to only important
species is inappropriate for the advisory concentrations.
Because advisory concentrations may be based on only three
species, the potential for missing important species is increased
greatly,
                           13

-------
4.2.3  Site-Specific Modifications

     The guidelines for deriving water quality advisories are
designed to produce a conservative value, thus/ advisories may be
over protective at many sites.  A procedure for site-specific
modifications, which could require generation of additional acute
and/or chronic data for species appropriate to the site, would not
only increase the data base for the chemical but would account
for water quality interactions which may affect the
bioavailability of the chemical.  A site specific modification
approach would be especially applicable to metals, organic
compounds that ionize, and those that have high partition
coefficients, and could also be adapted from the existing
procedure for site-specific modification of water quality
criteria.

4.3  Considerations for Developing. Other Advisories

4,3.1  Guidelines for Specific Chemical Groups versus Generic
         Guidelines

     The Subcommittee endorses the development of guidelines for
specific chemical groups, such as the guidelines reviewed by the
Subcommittee for low molecular weight, non-ionizable organics.
The Subcommittee agrees with EPA that, because of varying fate
and effects among groups of chemicals, initial development of
generic guidelines is inappropriate.  Guidelines  for specific
groups of chemicals should clearly indicate the chemical group
addressed in the document title, as well as in supporting text.

     The Subcommittee suspects that differences in guidelines
between most different groups of chemicals will be small.  The
acute to chronic ratio (ACR) of 25 used  in the guidelines may
serve for a broad range of chemicals, since a variety of
chemicals were used to derive it.  Similarly, the factors used to
calculate the advisory acute value (AAV) are related to the
number of Genus Mean Acute Values (GMAV) available, and should

                           14

-------
not be significantly different among chemical groups.  The use of
these parameters generically for chemical groups may be
substantiated by clarifying and justifying their derivation in
supporting documentation.

     The major differences between guidelines for different
chemicals will likely be in numbers and kinds of species to be
tested.  The AAV for ionizable molecules, metals and perhaps
other groups of chemicals may need to be based on a larger number
of representative species from both fresh and salt water.  Also,
water quality characteristics such as hardness and pH may need to
be considered.  If there is potential for a greater effect on
aquatic plants than animals, as with herbicidal chemicals, then
at least one acute test with plants must be required.

4.3.2  Development of Advisories for Wildlife

     The Subcommittee points out the fact that waterfowl, birds,
mammals, reptiles, and amphibians are valued environmental
resources and that they are exposed to chemicals via food, water
and other routes.  Water quality criteria incorporating
bioaccumulation, whether based on risk assessment assumptions or
FDA action levels for human consumption of fish, may not be
adequate for wildlife protection.  The reasons include important
differences in metabolism, feeding requirements and body weight.
For these reasons, the Subcommittee recommends that the EPA
consider the development of water quality advisories for
wildlife.
                           15

-------
5.0  HEALTH EFFECTS:  SPECIFIC ISSUES

5.1  Data Base Development

     The basic methodology for risk assessment outlined in the
"Guidelines for the Preparation of Office of Water Health
Advisories" (Appendix C) do not differ markedly from those
currently used by the Agency for setting drinking water health
advisories.  Current methods for developing health advisories to
protect human health are based largely upon an intensive
collection and review of the existing literature.  Research needs
may be identified, but, unlike establishment of criteria for
protecting aquatic life, additional laboratory research is almost
never conducted.

5.1.1  Secondary Sources

     The guidelines call for the use of secondary sources, which
summarize, annotate and compile primary or journal literature, in
preparing health advisories.  While the Agency should use the
existing literature efficiently, misconceptions and confusion can
arise for the use of secondary sources.  Misconceptions can be
introduced due to the fact that the studies likely to be
summarised in secondary sources were originally designed for
purposed other than safety evaluation.  Confusion arises because
the details necessary for appropriate integration, understanding
and analysis of the data are often omitted in such summaries.
Although they may serve as efficient directories to primary
literature, the exclusive use of secondary sources can not be
supported by the Subcommittee.

5.1.2  Literature Searches

     The Subcommittee considers the restriction of the literature
search to the most recent 5 years, as prescribed in the guide-
lines, to be a major problem.  Such a restriction discourages and
may prevent the use of many older studies that report significant

                           16

-------
exposures and effects in humans, especially as a result of
industrial exposures.  These studies are often supported by
experimental studies in animals which provide basic information
on acceptable levels of industrial exposure in humans,   While
concise presentations may be appropriate, responsible development
of a risk assessment requires thorough collection,  review and
critical interpretation of all pertinent data.

5.2  Modify, ing Factors

     The proposed use of "modifying factors" appears to the
Subcommittee as a major methodological feature that
differentiates the advisory-setting procedure from other risk
assessment procedures previously reviewed by the Science Advisory
Board,  As presented in the guidelines, the purpose and necessity
for this feature is unclear.  The presently used methodology for
establishing water quality criteria encourages the use of
scientific judgment in the selection of the most appropriate data
for assessing risk and uses four uncertainty factors which can
vary depending on the quality of the data.  Data on
pharmacokinetics and bioavailability are also used quantitatively
to adjust data in risk assessments.  However, the proposed
modifying factor appears to be little more than a "fudge factor"
which has a high potential for abuse in "adjusting" the outcome
to preconceived values, rather than values that are consistent
with scientific and toxicologic data.  In the opinion of the
Subcommittee, the proposal to use modifying factors is not well
developed, is not necessary, and can therefore not be endorsed.

5,3  Data Analyses

     Since the proposed guideline documents are similar in purpose
and methodology to water quality criteria documents, it is
reasonable that the outline and topics included therein are also
similar.  However, the tone and tenor of the guidelines reflects
a substantial degradation of the risk assessment process.  The
document discourages the critical analysis and synthesis of data

                           17

-------
where it should be encouraged.  The guidelines call for an
encyclopedic presentation of facts in a rigid format, followed by
a risk assessment protocol which emphasizes arithmetic rather
than analysis.

     The Subcommittee encourages guideline revisions that promote
standardization within sections to encourage comparison between
studies.  In addition, it recommends that discussions be included
to relate the elements of the outline to the derivation of
advisory concentrations.

5.4  Conclusion

     Most of the criticism raised by the Subcommittee can be
related to a fundamental flaw in the basic assumptions underlying
the development and use of water quality advisories to protect
health.  Namely, this assumption is that a good, short risk
assessment is substantially easier, quicker, and less expensive
to prepare than are long reviews and risk assessments.  While a
properly prepared and appropriately concise health advisory could
save paper, it will reguire the same level of analysis and effort
needed for a criteria document.  In many respects, the imposition
of an economy of words requires more effort in preparation than
longer detailed reviews.  In fact, the advisory process used for
health effects assessments may compromise the Agency's efforts to
provide sound guidance on tolerable levels of compounds in
ambient water.
                           18

-------
      APPENDIX
Request for the Review

-------
            UNITED STATES ENVIRONMENTAL PROTECTION AGENCY
                           WASHINGTON, D.C, 20460
                             30 BW                     „„,„„
                                                          WATER
MEMORANDUM

SUBJECT;     Request  for SAB Review of Water Quality Advisory
             Guidelines

FROM:        Edmund M. Notzon, Director fcf.*^***^-*/ /Wfj^T^
             Criteria and Standards Division (WH-S85)

TO:          Terry Yosie, Director
             Science Advisory Board (&-1Q1)


     I an requesting the Science Advisory Board review draft
guidelines developed for preparation of water quality advisories
for both human health and aquatic life protection.  The draft
guidelines are attached as are copies of completed advisories
which were developed using the draft guidelines.

     Water quality advisories are intended to be used as a
supplement to our efforts to develop water quality criteria
recommendations under Section 304(a), of the Clean Water Act.
Advisories are designed to fill the gap between the large number
of pollutants and the limited number of criteria documents we
are able to produce.  Advisories basically provide the Agency's
best scientific judgment applied to existing information,

     Please let me know as soon as possible when the Science
Advisory Board will be able to conduct a review of the advisory
guidelines.  Dr. Frank Gostomafci (475-7321) may be contacted
for further information.
Attachment*

-------
                 APPENDIX B
Guidelines ....for Deriving Ambient Aquatic  Life





                    Concentrations

-------
                                        :i/£v'
GUIDELINES FOR  DERIVING AMBIENT AQUATIC LIFE
           ADVISORY CONCENTRATIONS
                                           ~mfl.fi
                                            - r
  Office of Water Regulations and Standards
       Criteria and Standards Division
              Waihington,  D.C.

     Office of Research and  Development
         Environmental  Research  Lab
                Duluth,  MN
         Environmental  Research  Lab
             Narraganaett, RI

                 June 1987

-------
I.    introUuct
         Aquatic life advisories  will  oe  issued  for  selected
         chemicals £or which not  enough toxicity,  bioaccustulation
         and/or field data are available  to  allow  derivation of
         ambient *ater quality criteria for  aquatic  life  using
         the procedures describe'! in "Guidelines  for Deriving
         •jj'im-arical National Water Quality Criteria £or the       . .
         Pro--Action of Aquatic Organises  and Their Uses"  (Stephen
         et 41. 1935), hereinafter referrred to  as the "National
         Guidelines".  Aquatic Life advisories will  contain
         compilations and interpretations of available data
         than, \cs lirectiy pertinent to the  derivation of
         aquatic life advisory concentrations.

         Aquatic life advisory concentrations are  intended  to
         be used mostly for evaluating the aquatic toxicity of
         concentrations of pollutants  in effluents and ambient
         waters, whereas water quality criteria for aquatic
         life provide a stronger  basis for regulating concentra-
         tions of pollutants in effluents and ambient waters.
         A-lvisory concentrations  have the following two intended
         uses:

         1.  Advisory concentrations are intended to be used to
             interpret data on concentrations of chemicals in
             ambient water.  If the concentration of a chemical*
             in ambient water is  equal to or below .the aquatic
             life advisory concentration for that chemical,
             there is probably no cause for concern about
             effects on aquatic organisms and their uses.  If,
             however, the ambient concentration is above the
             advisory concentration, the discharger should
             quicKly evaluate the available exposure and effect
             data to determine whether it is prudent to:

             a,  obtain additional data concerning  the concen-
                 tration of the chemical in the effluent and/or
                 ambient water;

             b.  obtain additional  laboratory and/or  field data
                 on the effect of  the  chemical on aquatic
                 organisms and their uses so that a  more accurate,
                 and usually higher, aquatic life advisory or  a
                 water -quality criterion can be derived;

             c.  conduct acute and/or  chronic toxicity testa on
                 the effluent; and

             d,  reduce the ambient concentration of the chemical
                               -2-

-------
    After a reasonable period of time,  the
   • regulatory agency should evaluate all available
    pertiiant data, concerning the aabient concentration
    and the effects of the pollutant on aquatic life
    to determine whether it is appropriate to take an/
    action such as establishing a permit limit Cor the
    pollutant or requiring toxicity tests on the
    effluent.  Such agency aiay choose to regulate
   . either before or after collecting additional data,

2,  Advisor/ concentrations are intended to be used to
    help the U.S. SPA select chemicals for which water
    quality criteria for aquatic life should be derived.
    Any chemical that is found to be present in a
    considerable number of ambient waters at concentrations
    simil-ar to or exceeding the advisory concentration
    may become a candidate Cor derivation of water
    quality criteria for aquatic life.  Thus advisories
    will provide dischargers with advance notice of
    chemicals for which criteria might be derived so
    that they can generate additional data that might
    be useful for revising the advisory concentration
    or for deriving water quality criteria for aquatic
    life.

Additional guidance on appropriate regulator/ uses
of advisory concentrations and criteria  should be
obtained from the Criteria, and Standards Division,
Office of Water Regulations and Standards/ U.S.  EPA,

The procedures described in the National Guidelines
will be used as much as possible in  the  derivation of
aquatic life advisory concentrations.  Whenever  a
procedure described in the National  Guidelines  cannot
"oe used (usually because some required data  are  not
available), a procedure that  (a) follows  as  closely  as
possible the procedures described in the National
Guidelines and (b) is compatible with the  intended
uses of advisory concentrations will be  developed  for
use in deriving advisory concentrations.   Aquatic life
advisory concentrations caa be based on  fewer  data
than can water quality criteria  for  aquatic  life
because advisory concentrations are  not  intended to
have as much regulatory  impact  as criteria.   However,
to be compatible with the  first  intended use,  advisory
concentrations must be derived  so as to ensure that
                      -3-

-------
          they are rarely, if ever,  higher  than what  the Criterion
          Continuous Concentration (CCC) -would be if  enough data
          were dvailable to allow derivation of a. national  aquatic
          criterion £or the chemical.  The data requirements and
          procedures use-1 £oc -iecwing aquatic life advisory
          concentration is rarely, if ever, above what the  CCC
          wouL'1 bs.  Thus, whenever i national criterion  is
          derived  for a pollutant foe which an advisory
          concentration is already available-,  the CCC will  almost
          always be higher than the advisory concentration.  On
          the other haaJ.  an advisory concentration  that  is too
          much lower than the CCC will cause unnecessary  concern
          about various chemicals, effluents and ambient  waters,
          To be most useful, the advisory concentration should
          never be above what the CCC would be and should rarely
          be more  than a factor of 10 less than the CCC.

      D,  in order to obtain acceptable advisory concentrations
          for the  least cost, the data requirements and procedures
          jsed for deriving aquatic life advisory concentrations
          will be different for different classes of chemicals.
          when possible, classes will be defined so that  data
          requirements and procedures can be appropriately based
          on the biological, chemical, physical and toxicological
          properties used to define the class.

II.   Low Molecular Weight Jlon-idnizable Organic Chemicals

      A.  This class of chemicals is  not very  well defined yet.
          It is expected,  however, that all low molecular weight
          non-ionizable organic chemicals will be in this class
          after an upper limit on molecular weight has been
          established.  It might be possible to expand this
          class to include a wider range of chemicals within
          certain  limits.

          I.  'This class ia intended  to be limited to chemicals
              for which ther« is no reason to  suspect  that  the
              range of acute or chronic sensitivities of saltwater
              species will differ substantially  fro» those  of
              freshwater species.  Therefore,  unless thare  is
              substantial evidence to the contrary,  the data
              available for freshwater and saltwater species
              should be considered together  in order  to derive
              an advisory concentration that will apply to both
              fresh and salt water*   Because of  the  differing
              ionic compositions of the waters,  it seems reasonable
              to assume that the  toxieities  and  BCFs of  organic
                               -4-

-------
        chemicals that ionize and inorganic chemicals  are
        Likely co (iiESec in Sresh and salt water.

    2,  This class I* Also intended to be liinitsd  to
        chemicals whose range o£ toxiciti^a to aquatic
        animal spscles id relatively snail, so that the
        requirement's  Sor acute values do not have  to
        include very many species and do not have  to be
        very specific.  Thus this class o£ chemicals should
        .lot include any pesticides that are intended to be
        effective against any aquatic or terrestrial
        animals or any metals.

    3-  This cla^g is also intended to be Limited  to
        chemicals that are not especially toxic to plants,
        so that tests with aquatic plants do not have to
        be requital.  Thus this class of chemicals should
        not include any herbicides.

B.  An aquatic life advisory concentration should not be
    calculated for a. chemical unless data ace available
    from acceptable acute tests *ith at least three aninal
    species, such that:

    1.  at least one species is a fish In the class
        Osteichthyes  in the phylum Chordata.

    2.  at least t*o species are invertebrates such that;

        a.  at least one species is in the class Crustacea
            in the phylum Arthropoda.

        Q.  the'other species is either  in the phylua
            Mollusca  (test with embryos "and larvae  leading
            to a 96 hour £CSO or LC50) or  in  a different
            family of the jphyluw Arthropoda.

    3,  at least one  ia a freshwater species.

    4.  at least ona  is a saltwater species.

         Available data from foreign  species  should bs
    included in the advisory* but not  utilized  to  derive
    an advisory recommendation  unless  other  required
    data is not sufficient.
                         — 5 —

-------
     Because .!iaiy o£ the chemicals in this class -ira highly
     volatile or .legradable, -acute tests with animals and
     tests with plants that *r-3 otherwise acceptable (in
     terms of acclimation, control mortality, ate. as
     described in the National Gvii del in.es) are acceptable
     •for  this class 06 chemicals only if:

    |j£.   For flow-through t^sts, the concentrations wars
    ','.".  measured.  If concentrations fluctuated unreasonably,
        •the test should not ba
     2.  Foe renewal tests t  th« organisms were exposed to
        fresh test solution a.t least once evety 24 hours
        an»1 either (a)  the  properties of the chemical
        indicate that its concentration in water should
        not decrease by more  than 501 in 24 hoars or (b)
        measurements on tests solutions showed that the
        concentration of  test material did not decrease by
        aote than 50% in  24 hours.
                                               +
     3.  For static tests, either  (a)  the properties of th«
        chemical indicate that its concentration in water
        should not decrease by more  than 50% in .96 hours*
        ("o) measurements  on teat solutions showed that 'th«
        concentration o£  test material did not decrease by
        more than 50% Iron  the beginning to the end of the
        test or  (c) results of a nominal or measured static,
        test should be  multiplied by  a factor obtained by
        dividing a f low-through 96-hr LC50 by a- comparable
        static 96-hr LCSO.  The comparable flow- through
        and static tests  must b* conducted on the chemical
        in the same laboratory using  the same water and
        organisms from  the  same sources.  The results of
        the flow-through  tests must be based on the time-
        weighted average  measured concentrations of test
        material and the  results of  the static test must
        be baaed on th* concentrations measured at  the
        beginning of the  test.

D.  Although data from  tests  with aquatic plants ar«
    desirable th«y are  not  required because  for many  chemicals
     it appears that aquatic plants  are  adequately protected
     if aquatic animals  are  adequately protected.

£.  For each species for  which at  least  one  acceptable
    acute value  is available, determine  a Species  Mean
    Acute Value  (SMAV)  using  the  procedure  described in


                          -6-

-------
     the National Guidelines-  (if data from tests in both
     fresh and  salt water are available for a species such
     as .-striped bass, all the data should be used together
    -*hen determining the SMAV for that species.)  Than
     calculate  a Genus Me-in teute Value (GMAV) £pr each
   '  genus for  which at  Isast one SMAV  is available.

F.;;||itB F-W should be calculated using the procedure
   '"'described  in the National Guidelines if GMAVs are       - -
     aviilAOle  Cor at least one animal species in at  least
     eignt different families, such  that either?

     1.  the acute data  requirements specified in the
        National Guidelines for either fresh or salt water
        are met, or

     2.  all the following ar« includeds

        a,  three families  in the phylum Chor«Sata such
            that:

            (1)  at least one species  is in  the family
                 Salmonidae.
            (2)  at least on* is a  freshwater species.
            (3)  at least one is a  saltwater species.

        b.  i  saltwater penaeid shrimp or mysid.

        c.  a  freshwater cladoceran.

        3,  a  family in a phylum other than  Cliordata or
            Arthropoda.

        e.  two other families not  in  the phylum Chordata.

         As described in the national  Guidelines,  in some
    situations a calculated FAV should be  lowered  to protect
    an important animal species.

G.  If th« requirements far calculating an FAV  are  not met,-
    calculate  an Advisory Acute Value  (AAV)  by dividing
    the lowest available GMAV by the appropriate  factor:

    Number of  GMAVs                         Factor

           3                                 11.0

           4                                 10.0
                          -7-

-------
     Number, of GMAVs

           5

           6
           3

           9 '

           10

           11

           12

           13

           14

           15

           16

           17

           18

           19

      20 or more
    The AAV is intended to be equal to or slightly below
    what th* FAV would be if one could be calculated.
    Since the factors for 8" GMAVs and above are only to be
    used when those GWAVs are not acceptable under the
    National Guidelines, the lowest factor has been set at
    2, to provide a conservative estimate for the advisory
    concentration,  if there are 3 acceptable GMAVs, then
    an PAV can be calculated directly.

H.  II three or more experimentally-determined acute-
    chronic ratios (ACE) which are acceptable based on the
                         -8-

-------
    National Guidelines ara available for the chemical,
    determine  the Final Acute-Chronic Ratio (FACR) using
    the procedure described in the* National Guidelines,
    If  £e*qr than three acceptable axp*cimentally-deterained
    ACRs are available, use enough assumed *CRs of 2S so
         the total n amber of exparimeatally-tleter mined and
            ACRs equals three (over 90% of the ACR reported
       both Kenaga (1932) and Call et al.  (1985) were'less
         2^5 and nearly ail the F&CRs used to derive watir
    quality criteria for aquatic life have been less than
    25),  Calculat-a the Advisory Acute-chronic Ratio (AACR)
    as the geometric mean of the three-*ACRs, Thus is no
    experimentally-determined acute-chronic ratios are
    available/ the AACR is 25.

I.  Calculate  the advisory concentration by dividing the
    FAV (or the AAV if an FAV cannot be determined by the
  -  FACR (or the AACR i£ an FACR cannot be determined).

J.  If necessary, the advisory concentration should be
    lowered to one-half of the lowest EC50 for an important
    aquatic plant species for which the ECSO is available
    from' *n acceptable test, based on the National
    Guidelines, in which the concentrations of teat
    material were measured and the effect was biologically
    important.
 i                                                          *
K.  If a Maximum Permissible Tissue Concentration  (either
    an FDA or othar regulatory action  level  far seafoods
    or from wildlife feeding studies, as described  in the
    National Guidelines) is available, bacfc-calculate to a
    concentration in water using a measured BCF (or  a
    predicted BCF if a measured BCF  is not  available).   If
    necessary, the advisory concentration  should be  lowered
    to be equal to the calculated  concentration.

L.  The advisory should be stated ass

         If the measured or estimated  ambient  concentration
         of (a) exceeds (b) in fresh or salt water,  one or
         mor* of th« following options must be completed
         as quickly as possible;

         1.  obtain additional data  concerning the concen-
             tration of (a) in the effluent and/or ambient
             water;

         2.  obtain additional laboratory and/or  field data
             on the effect of  (a)  on aquatic organisms and
             their uses so that a  new  aquatic life advisory
             or a water quality criterion can be derived;
                          -9-

-------
         3.  conduct acate and/or chronic toxicity tests
             on the affluent?

         4.  rsduce the concentration.

         After a reasonable period of time, unless a
         of all avail-able data concerning the ambient concen-
            and the effects of (a) on aquatic life demonstrate
    .t'aat the ambient concentration is low enough,  it fnust be- -
    reduced,

    '.vhace (&}j~ insert name of chemicai and

          (b) = insert advisory concentration

M.  Caveats should be added to the advisory statenent in some
    ^situations;

    1,  if data for a commercially or recraattonally
        important species indicate that the species might
        not be adequately protected by the advisory
        concentration, but the data do not justify lowering
        the advisory concentration (for exanple, because
        the concentration of test material wer* not
        maasucsl),  caveat should be added stating that tfc*
        species might not be adequately protected.

    2,  If SCSOs for a variety of species of algae (or
        aquatic plants in general) are below the  advisory
        concerttration, a caveat should be added stating
        that algae (or aquatic plants) might not  be
        adequately protected. •
                         -10-

-------
                           References
Call, D..JT.. * L.T. Brooke, M,L. Knuth, S»H.  Poiciac and M.D.
               1,995.  Fish subchroaic toxic it/ prediction
           for industrial organic: chemicals  that produce
           id,  Snviron. ToxicoL. Chem. 4t335-341.

Kenaga,  E.£.  1932.  Predlcatability o£ chroaic  toxicity  from
     acate toxicity of cheiticals in fish and  invertebrates.
     Environ. Toxicol. Chen.  li347-358.

Stephen, C.E./ D»I. Mount, O.J. Hansen* J.H. <3entilef Q.A.
     Chapman and W,A. irunga,   1985,  Gaidellnes  Cor deriving
     numerical national water quality criteria  for  the  protection
     of. aquatic organisms and their uses.  PB85~227049»  National
     Technical Inforaation Service, Springfield* Va*
                               -11-

-------
   f*       '         APPENDIX C
   ,:-"^:,~ •;





Guidelines  for  til® Preparation fif Office of Mater
                                       t

                 Health Advisories

-------
                   APPENDIX  C
Guide_lines_for the Preparation  of  Office Qf_wa£.eg





                Health Advisories

-------
GUIDELINES FOR THE PREPARATION OF  OFFICE  OF  MATER
                HEALTH ADVISORIES
      U.S. Environmental Protection Agency
  Envlronmtntal Criteria and Assessment Office
              Cincinnati, OH  41268
                   May 6, 1987

-------
 INTRODUCTION

 These guideline! were prepared to assist authors and others  involved \n the
 preparation of  Health Advisory |HA) documents for the U.S. Environmental
 Protection Agency  (EPA),  Office of Orinklnf Mater (QDW)  and the  Office of
 Water Regulations  and Standards (OURS),

 The HA document provides a quick review and summarization of the  key
 11 teraturf^en the  physical, chemical and toxUologlcal properties of the
 specif led^itflilcal.  Kty literature contains Information  hawing a direct
 effect on .in* estimation of an MA or provides Basic scientific  Information  on
 the chemical 0f interest.  When available, existing documents and reports will
 be  heavily relied  upon  to complete this document.  The HA d§cument is  used  by
 £PA to aid In determining control priorities when tht chemical  1s present  In
 ambient  or drinking water.  The calculated and extrapolated  HA  values  included
 within the HA document  are not legally enforceable ambient or drinking water
 standards.
 LITERATURE  SEARCH

 To  provide  literature search and acquisition support consistent with the
 magnitude of  the entire effort, the literature search for a HA document will
 rely heavily  upon those searches already performed In support of available,
 good quality  primary references or summary documents on the chemical.  As a
 minimum, a  limited  (last five years or 2-3 years prior to most recent
 reference dted in  an existing document) automated literature search will be
 performed to  ensurt Identification of tht mort .recent technical IHeraturt,
 Typically,  the available summary documents and IHtraturt starch outputs wrtli
 then bt forwarded to tht author for his selection of tht key primary
 literature  rtferences to bt acquired during preparation of tht HA.  On
 occasion, an  Initial selection of the key literature will bt performed by EPA
 staff prior to the  author's Involvement.  The contractor's support staff win
 acquire coplts of the selected reftrtnces and thtst will be forwarded to the
 author to permit Initiation of the original writing tasks.

 Primary References                    '  -

 Due to the  magnitude of tht effort, existing documents and reports must  be
 heavily relied upon.  To avoid transferring Interpretations! or typographical
 errors to the HA from secondary references spot checks will bt done.
 Secondary references (e.g., chemistry texts) may be used tn the preparation  df
 Section II  (Gtntral Information and Properties).  Secondary rtftrtnces  will  be
 cited when  tnty art tht original sourct of Information.  For example,  an
 existing IPA  document My bt a secondary reference  for toxicology  data but  the
 primary soyrct for  txtrapolations from that data to hymans  (i.e.,  primary
 reference for existing guidelines or standards).

 Kev Reftrtnct

Only a limited number (average of 20) of literature rtftrenets art usually
 specifically  dted  In the HA document.   Tht selection  of  reftrencts to be
                                       -2-

-------
 - '- - - '• ..  H« «••'. i.',/  -•"•*  M.WJL  *.•'','vj*  ilcjj  '. i"i t i IB pf epdf <311 on process.   The  HA
 documents are  not  comprehensive  reviews of all of the literature on each
 ambient  or drinking  water contaminant,  Conciseness Is maintained by selecting
 only  the  most  Important/valid  references  for citation.  Emphasis must be  on
 the determination  of the NOAEl and/or  the LQAEL.  This establishes the overall
 scope and technical  emphasis of  the  HA.

 The selection  of key references  Is also Influenced by the technical "emphasis"
 °"  oral exposure.  Preferably, studies selected for Inclyslon into 'the HA win
 Involve  ('If'oral exposure-food,  drinking  water, gavage. (2) .dermal
 exposure/ibsorpnon  and  (3)  Inhalation exposure If pertinent Information on
 the chemlcaVls provided or  if there ar*  no other data available.  Studies
 involving other  typts  of exposure are  only useful when more pertinent data are
 licking.
                      *                              y
 Pesticides/Confidential Business Information

 For HA documents on  pesticides,  etc.,  a special consideration is often
 encountered.   Much of  the toxicology and  pharmacology data may hive been
 developed by industry  and,  for proprietary business reasons, classified as
 Confidential Business  Information (Cil).  This CBI  Information cannot be
 released  to the public unless  specific authorization  U provided by  the
 submitting Industry.   When  such  a situation 1s encountered, tilt document will
 use the CBI information  1n  calculating the advisory,  and non-CBl summaries
 will  be included.  Sanitized copies  of any literature that  Is of  importance
 should be on file.   The author will  be authorized  to  access the CBI and
 Instructed to  Incorporate non-CBI summaries Into  the  HA from the CSI  sources.
 Here  a disclaimer  Is employed  to explain  why certain  details (CBI)  are not
 included.   Special Instructions  are  provided  to  the author  regarding  the*
 preparation of non-CII summaries for Inclusion  In  the HA document.

 Reference Hard Copies

 Two copies  of every  cited reference  will  be obtained  and submitted to EPA.
 The author may dte  references from  his/her personal  reference  collection or
 other readily accessible sources.  It  1s  emphasized  that two legible copies  of
 each  reference must  be provided  with the  HA document,"draft.

 Translations

 Due to the  significant costs that can  be  Incurred,  translations of foreign
 language  references  art generally avoided and,  if  absolutely needed, are dealt
with on i  case-by-case basis.  The author must  identify all desired
 translation requirements early In the  HA  preparation process and check with
 the IPA HA manager to  see If IPA already  possesses a  translation or If IPA
chooses to  translate the reference  Itself.
DOCUMENT ORGANIZATION

The organization of an  HA  document Is designed to achieve a uniform and
effective summary of the pertinent data needed to determine a HA.  An outline
                                        ,3,

-------
 of  the major  sections  1s given  later Including suggested contents  of each
 sections,   More  information  1s.  described here concerning the technical
 emphasis,  presentation  format,  and document tone desired for the document.

 This  rigid structure  frequently poses organizational problems with respect  to
 presenting the results  of experimental studies, since data pertinent  to  two or
 more  sections of  the  outline are often presented 1n a single research  paper.
 Experience;-Jias shown  that the best way to solve this problem is to "summarize"
 the resylffeof each study, presenting the results of Individual experlnents
 {or groypf^f similar experiments) under the appropriate outline headings.
 This  plan  dors lead to  some  redundance with regard to the description  of
 experimental conditions  (see below), byt nevertheless it Is the format that  is
 most  useful to tht various users of the HA documents.
                                                    .*

 TECHNICAL  EMPHASIS

 The HA document must  be  a technical docyraent.  EPA science policy, economic or
 political  considerations are not an issue during the selection of references
 or the preparation of the document.  The technical emphasis in the HA document
 Is with  the presence  of  the  specified chemical In water.  Thus, all chemistry
 related  portions  of the  document (physical, chemical properties, chemical
 analysis)  should  emphasize the  chemical's properties and behavior in water and
 accumulation In aquatic  life.   The health effects portions of  the document
 should emphasize  the  compound's lexicological  properties by the most relevant
 route  of exposure, (I.e., oral  and derrwal exposure).  Toxicology  Information
 for other  routes  of exposure may be Included as background material for
 toxlcologlcal discussion.  However, due to the .limited  scope  of  these
 documents  and magnitude  of the  effort, such discussion  will tie minimal  should
 H occur at all.


 FORMAT  FOR  PRESENTING EXPERIMENTAL RESULTS

 Each experiment (or group of related experiments)  should  be described  1n a
 paragraph  that is an  independent unit, able to stand alone.   This  format  H
 needed  to  provide a concise  summary of the key information  and for  locating
 Hems  of Interest relative to a specific exposure  situation.   The format  is
 also useful during HA revisions yhen additions, deletions  or  rearrangements
may be made.

 The text describing the  results of  Important  experimental  studies must provide
 sufficient  experimental  detail  to allow the reader to  form an independent
 opinion  regarding tn« quality and  importance  of  the results.   (Such opinions
may. be offered by the author of the HA but must  be clearly Identified as
 such).   If  available, the following  Information must be Included (more or less
 in order)  \R the  paragraph describing an experiment or series of related
experiments.

    1,   The literature  citation  (in  proper  format) is contained in a brief
         Introductory sentence.  The  format  for  citing references *t1ll be
         provided and examples  are  given  later.
                                        -4-

-------
     2,   Species of test  animals  {ana  strain, sex, age. body weight   \f
          critical  to tne  results  interpretation),

     3.   Chemical  form (e,g,,  "copper  as
              
-------
      ,  calculate  the  average daily aose (mg/kg/day} over  the duration of the
 study.   In calculating  tht dost in mi/kg/day, carry only  as many  significant
 figures as are  reported  by the authors. Often, all Information required  For
 calculating the dose  in  «g/kg units 1s not available in the cited reference.
 and assumptions must  tie  made.  AH such assumptions must  be clearly  laoeled'as
 such.   In order to  keep  track of all experimental details and tnt dose
 calculations.  It  Is suggested that a worksheet be filled  out for  each
 llterature.;Ftport used,  providing tht basis for all details and mathematical
 conversion"^-made.   These worksheets, as well as other supportive materials,
 should  become part  of tht document file and a copy provided to EPA,


 TON!
                      /                              .•
 All  sections of the HA document must be concise, accurate, factual,  objective
 and neutral  summaries of the Important work: that has been reported in the
 literature.  The author  of the' HA document is not permitted to make deductions
 or  extrapolations from  the data into the ftport, regardless of the logic and
 accuracy  of  the extrapolation.  Deductions and extrapolations made by the
 author  of  the citation may be reported, but must be clearly indicated as
 such.   It  1$ emphaslied  that the HA author's technical expertise and Judgement
 is  required  to  make critical decisions during the selection of those key
 references  to be reviewed and cited 1n the HA.  8ut, with  tht exception of the
 section on  "Quantification of ToxUologlCIl Effects* the HA document 1s to be
 only a  summary  (not an extensive critical review) of tht key  Information.

 Example text

 The  following paragraphs are offered as examples of  the organization, content
 and  style  that  1s desired In descriptions of experimental  results.

     Single doses of compound X at 1.0, 2.5, 5 and  10 mg/kg administered by
     intravenous injection in (vehicle) to rats  (age, sex*  strain), 10 animals
    per dose, were  shown "to cause (effect) it doses  >2.§ mg/kg.   A dose of  1.0
    mg/kg caused no (effect) (Hllltr et al., 1S83).

    Adult Sprague-Dawley male rats were administered compound X  In their
    drinking water at dosts of 0, 200 or 2,000  ppm (0, 10  or  100 mg/kg/day)
     for U consecutive days (Smith, 1977b).  The  number  of animals used was
    not reported,  A NOAEL of 200 ppffl dost  (10  fltg/kg/day}  was Identified
    showing  no  effect(s) on 	.  The 2,000 ppm dose  (100 mg/kg/day)  caused
     (effect).   In a liter review of tht study,  U.S.  EPA  (19SO) Identified
    several  problems with tht study, which may  have compromised  tht  results.

The format specifications should be approached  as  the  maximum requirement for
each summary.   Typically, sont abbreviation  is  acceptable  to make the  HA
concise and  describe  those aspects of each  study that  contribute the important
 information,  Tht author, however, should be  conservative in preparing
abbreviated  summaries.   The IPA reviewers have  final authority over  HA
document content.  The deletion of "unnecessary" experimental details  Is
easier than  later revaluation of the  primary  reference  to tipand tht text.
                                        -6-

-------
     The section  on  "Quantification of To*1cologtcal Effects",  However, does
         tnat  the author mike scientific Judgments and thus,  departs  from the
 objective,  neutral  and factual summary tone of all other HA  document
 sections.   In this  section, the author must select specific  sets  of  data to &e
 used for  each _-WA calculation.  This data is then used with EPA established
 guidelines  to estimate acceptable levels.


 HEALTH  AO^ISORY  CALCULATIONS
          , -*>
          . %•%-
 Calculation's "of,  HA  values are required for 1-day and 10-day  periods  as well  as
 longer-term and  lifetime HA values 1f adequate subchronlc and chronic toxlcity
 data are  available.   For all such calculations, extrapolation of  animal
 toxlelty  data to human* will probably be required.  The 1-day, 10-day and
 longer-term for  a child HA calculation shall be based on a 10-kg  child who
 consumes  one  IHer  of water per day.  Tht adult longer-term and lifetime
 health  calculations  shall oe based on a 70-kg adult who consumes  two liters  of
 water per day.   lach  HA calculation requires the selection of the best single
 data-set.   These data must be (1) of high quality, (2) from a study where
 evaluations of target organ effects have been observed. (3)  from a study using
 the  most  relevant route of exposure and (4) for a  study duration comparable to
 that  for  the  HA  value being calculated.  A 1-day HA can use up. to a 7-day
 study;  a  10-day  HA  can use up to a 30-day study; a longer-terra HA can use  a 30
 to <90-day  study or  10% of the animals' lifetime;  a lifetime HA can use >90
 day  study.  Appropriate uncertainty factors are then applied to derive the
 calculated  HA exposure levels discussed below.

         Quality
The quality of available data-sets will be a Judgment resting primarily with*
the HA author.  Publication of  tox1c1ty data In  reputable  "peer -reviewed*
toxicology or medical Journals  should be  used is one measure of data quality.
Informal coordination with the  study authors or  other experts active in this
technical field may also-be required to assist  In assessing the quality of
data being considered as the basis for an HA calculation.  Discussion with  the
EPA staff is encouraged during  the course of the assignment to review the
quality aspects of data being considered.

Target Organ iffects Data

The selected data-sets must be  from studies where observations for  "target
organ effects" (e.g., observations for effects  other  than  or  in addition  to
lethality) have been made.  The following priorities  should be applied  in  the
selection of studies for use in calculating HA  values  (1)  dose-response
studies in wMcn a NOAEL 1s Identified,  (2} studies employing a  single  dose or
multiple dost that did not produce an effect  (a mini mum NOAEL) and  (3)
dose-respoflse studies In which  the  lowest dose  tested still produced an effect
(LOAEL only).  In cases where there ire  two or  more  sets of reliable animal
toxlcHy data* the data set that represents  the highest NOAH or  the lowest
LOAEL will be used,
                                        -7-

-------
 Lethality  data  (regardless of quality) should not be used as  the  basis  for 
-------
 Uncertainty	rac.to.rs

 A  NGAEt  or LOAEL  1s determined from animal toxldty data  or  human  effects
 data,  for animal data, this level 1s modified by an uncertainty  factor mainly
 because  tnerejls  no universally acceptable quantitative method  to  extrapolate
 from animals 'to humans,  Tne possibility must be considered  that  hymens are
 more sensitive to the  toxic effects of chemicals than are animals.   For human
 data, anjncertalnty factor 1s also used to account for the  heterogeneity  of
 thf huma^populatlbn In which persons could exhibit differing sensitivity  to
 toxic chepfGals.  The  suggested modification of the guidelines  set forth oy
 the National'Academy of Sciences and modified by the U.S. EPA I986a typically  .
 used 1n  establishing uncertainty factors are as follows;

 Standard Uncertainty/Factors (Ufs)

    «    Use a 10-fold factor when extrapolating from valid  experimental
         results  from  studies using prolonged exposure to average healthy
         humans.  This factor is Intended to account for  the variation in
         sensitivity among the members of the human population,  [1QHJ

         Use an additional 10-fold factor when extrapolating from valid
         results  of long-term stymies on experimental animal* when results of
         studies of human exposure are not available or are 'inadequate,  This
         factor is Intended to account for tne uncertainty In extrapolating
         animal data to the cast of humans.  [IDA]

         Use an additional 10-fold factor when extrapolating from  less than
         chronic results on experimental animals when  there  H no  useful
         long-term human data.  Tftls factor  1s* Intended  to account for the
         uncertainty in extrapolating from less  than chronic MQAELs to chronic
         NQAIis.  [105]

         Use an additional 10-fold factor when deriving  an RfO from a IQAEL
         instead of a NOAIL,  This factor 1s Intended  to account  for  the
         uncertainty In extrapolating from IQAils  to liQMls.   [lOL]

Modifying Factor  {«)

         Use professional Judgment to determine  another  uncertainty factor
         (Mf) that is greater than zero and  less  than  or equal to 10.  The
         magnitude of  the NF depends upon the  professional  assessment of
         scientific uncertainties of the  study and data  base not  explicitly
         treated at>o¥t, e.g., the completeness of  the  overall  data case and
         the nuiaoer of species tested.  The  default valye for  the w  Hi,

If the investigator feels that a modifying  factor  should be used, he  is
instructed to get in touch *Hh the appropriate  IP* contact for  clarification.
                                        -9-

-------
                                   REFERENCES

References elttd 1n the document are to be complete and follow a  standard
format.  When citing primary data given In a secondary or review document  the
original citation must be given along *1th the secondary source (I.e..  Stair
et al., 1978,-as cited Mi U.S. EPA, 1986).  The appropriate format for  the
citation tn the olbllograpny H as follows:

iansal, Q •«.£-. 19S3.   Adsorption  of  oxamyl  and  dlmtcron In «»ontmorm§n1te
suspensions.  Son  Sci, Sac. Am. J. 47:887-882.

U.S.  £PA»  "1969.  Thirteen week feeding study - dog.  Project 1210-239, Ace.
i§69!0.  OctoDer, Office of Toxic Substances, Washington, 0,C.

U.S.  EPA.   1986. Guidelines for carcinogen risk assessment.  Federal Register
5U185):339I2-34003,
                                       ,10-

-------
                                Office of *ater
                    Aqyatic  Life and  Hyman Health Advisories

 I ,    Executive Sunmary

 II.   Introduction

 III.  General Information

      A.  eStemkal and Physical  Properties
      B.  Occurrence
      C.  Environmental Fate
      Q.  Analytical Methods
      I,  Treatment    .'                             <
      f ,  Human Exposure

 IV.   Aquitlc Toxldty

      A.  Introduction
      8.  Freshwater Toxlclty
      C.  Saltwater ToxlcHy
      0.  Quantification of Toxlcologlcal  Effects
      £.  Selected Data Table
V.   Manflillan ToxUHy

     A.  Pharmacoklnttlcs

         1.   A6$orpt1on
         ?.   Distribution
         3.   Metabolism
         4.   Ixcretlon (f 11ml nit ton)

     8,  Health Effects -

         1,   Short-Ttrro ixposure
         2.   Oernwl/Ocylif Efftct?
         3.   longer -Ttrm Exposure
         4.   Reproductive Effects
         5.   Developmental Effects
         b.   Hutagenlclty
         7,   CarclnogenlcHy
         8.   Other Effects

VI.  Quantification of ToxlcolofUal Effects

     A,  Htmn Health
         1,   water Exposure
              a,   1-day HA (for drinking water)
              b.   10-diy MA (for drinking water)
              c.   longer-term HA (for drinking water)
              d.   Lifetime HA (for drinking water, fish and drinking water,
                   fish)
              e.   Cancer Risk Assessment (for drinking water, fish and
                   drinking water, fish)

-------
Vll.  Other  Criteria  and Standards (Including organolepUc data;

V1ILEPA Contacts

         1.   .Aquatic Life Advisories
         2.   ' OMnMng Wattr Advisories

IX.   References

-------
 I.    EXECUTIVE

 It.   GENEiAl  INTRODUCTION
      [This  section of the HA document Is standard  IPA  "boilerplate" that is
 Included  verbatim In all such documents.  Normally,  a  typed  version of this
 section will be  Included in the initial package to the author.]
               h Advisories that have betn derived In  the  document should
 summarlietf'here as presented in the OWS format]
                                                     contaminant  and a concise
                                                     general,  the emphasis of
                                                          of  various water
III. GENERAL INFORMATION

     [This section provides  an identification of  the
summary of Us physical  and  chemical  properties.   In
this section Is on the Identification or  characterization
contaminants.  It should Include the  following  information:]

Chemical and Physical  Properties

     [A list of fundamental  physical  and  chemical  properties that may be
useful  1n assessing hazards  associated with  contaminants of Interest.]
         CAS f
         CAS name
         Chemical formula
         Chemical Structure
         Molecular weight
         Physical state (at 25 degrees C)
     *   Melting point
     *   Soiling point
     *   Vapor pressure (2S degrees C)
     *   Specific gravity (25 degrees C)
     *   Water «olubiiUy (2$ deo/tts C)
     *   Octanol/water partition coefficient (log Kow)
     *   Taste threshold (water)
     *   Odor threshold (water)
     *   Odor threshold (air)
     *   8CF (range of 8CF for Illustrative purposes only)

Synonyms

     [A list of related names that are applied  to tht chemical of  interest.]

Occurrence

     (A brlif dtscMptlon of tftt common  uses and application  of  the  substances
of inttrtit along with an Indication of  th* amount  of  the substance  that  the
public could bt exposed to, especially those that Itad  to contamination of
ambient and drinking water.]
                                       -13-

-------
 Environmental  Fate and Human Exposure

 [A  brief  description summarizing the current Information  on  the  transport,
 fate,  and distribution of the substance 1ft the environment.   This section'
 should allow * relative source contribution determination,]

 IV.  AQUATIC TOXICITY

     This "4ftt1on will be provided by the EPA,

 V.   PHAftMACQklNlTICS

     [This section should be brief {even If the technical literature  is
 extensive In this area).  The pharmaeoklnetles data may provide  supporting
 evidence  of the quality or relevance of the selected toxlclty data.   This
 section Is divided Into fo«r subheadings, the contents of which  are detailed
 below.]

 Absorption

     [This section should present data that provide a quantitative  estimate  of
 the fraction of an oral dose that Is absorbed from the G! traet  Into  the body
 and for Inhalation, if the chemical 1s volatile, and dermal, even If  the
 chemical  Is or 1s not absorbed through the skin.  Data regarding the  time
 course of  absorption should be presented, but should not be confused  with the
 contents  of the section on excretion.]

 Distribution

     [This section should provide data regarding tissue uptake of the
 substance, with special reference to the question of whether any preferential
accumulation In a target organ occurs.  To this end, results expressed as
 (amount substance)/g tissue) are preferred to those that express results as
 total amount of substance In each tissue, although both are useful.]

Metabolism

     (This section should focus on covalent reactions  that  the  substance
undergoes  in the body, with special enzymatic reactions  (activation,
conjugation, etc.) and non-eniymatlc ones  (oxidation,  hydrolysis, etc.).
Where relevant, noncovalent reactions  (binding, adsorption)  should also be
covered, ]

Excretjo_n

     [Thlt stctlon should provide quantitative  data  regarding the  percent  of  a
dose that  Is eacreted from the body via  feces,  urine,  bile,  expired  air or
sweat.   Half-lift data need to be  included in appropriate sections  to provide
sense of  'residence tides" with continuous  or Intermittent  exposures.]
                                       -14-

-------
 Vl. HEALTH EFFECTS

      [This section Is  the west  Important part of the document, since H  is
 from these studies that  tht  quantitative assessment of risk (section V)  Is
 made.   Since *any  substances  show cumulative effects, studies {or individual
 experiments from more  extensive  studies) are organized according to the
 duration of exposure.   Each  study discussed should bt well explained providing
 Information on  dose, exposure concentration, duration and route of exposure
 and spedl* differences  that  are seen  in the available results.]

      {Human and animal  dlta  shall be described  In separate sections with the  -
 human  data presented first,   Additional subheadings should be Included in tnls
 section of the  document  to discuss  the available Information on systemic
 toxldty {as »  function  of short and long-term  exposure time)
 reproductive/developmental toxlclty, teratogenlclty, mMtagen1.c1ty, and
 carclnoftnicKy.   The  contents of these headings and subheadings are discussed
 below.   When there are  no data concerning any of these toxic effects, a
 general statement  should be made to that effect.]

 Human

     (Much of the  data  on human  exposure are derived from clinical case
 studies and epldemlologle studies.  These studies should be briefly summarized
 with critical effects  Identified.J

 Short-Term Exposure (Acute Exposure^   •

     [Studies involving  a single acute exposure or multiple exposure up  to 30
 days are  Included  In this section.  Adverse effects  on various  tissues and
 organ systems (e.g., hepatic  effects,  renal effects, etc.) are  presented,) *

 Long-Term  Exposure (Supchronlc and  Chronic Exposures!

     [Studies presented  ^1n this  section  involve exposure  periods  In  excess of
 30 days.   Effects  reported are  those other than reproductive/developmental.
mutagenic  and carcinogenic responses.  If there 1s evidence  for or  against
occurrence  of these endpolnts.  they should be described  under  the specific
subheadings.

Animals
Short-Term Exposure  (J*cut_e  Exposure)

     [Studies Involving a single  acute  exposure (Including LD$QS,
etc.) or multiple exposures  up  to 30  days  are presented.   Studies
demonstrating adverse effects on  tissues or  organ systems (e.g., hepatic
effects, filial effects, etc.) is  well as  Irritant properties are presented.
Data presented In these studies (other  than  lethality) serve as the basis for
deriving 1-day and 10-day HAs.]
                                       -15-

-------
 Dermal/Ocular  Exposure

      [This  section  should  contain descriptions of data on eye and s*in
 Irritation  and shin sensltUatlon,  If available.  Expectation of systemic
 toxlclty following  dermal  exposure,  if different only by degree, should  &e
 addressed in "the  section on  Pharmacotnnetles:  Absorption.]

 longer-Term Ixposure (SubchrpnU and Chronic Ixposures)

      [Studies  presented  in this section  Involve long-term exposure (from 30
 days  up  to-It  months) or lifetime (2* months for rats or mice).  Effects
 reported are t"hose  organ system effects  other than reproductive,
 developmental,  mutagenlc and carcinogenic responses,  fyrther, the effect that
 1s  tht most appropriate as an  Index  of chronic toxtcUy should be identified.
 Data  presented 1n the studies  included In this section will usually be used in
 deriving the longer-term and lifetime HAS.]

 Reproductive Effects

      [This  section  should  contain data describing the effects of the substance
 on  the reproductive success  of exposed parents and on the survival of
 offspring.  This  section should also Include data regarding any  Increased
 frequency in structural anomalies 1n offspring.  Scientific guidelines to be
 followed are given  In FR Sl{185);34027.]

 HutagenlcUy

      [This section  should  contain the results of experiments  designed to
 assess the mutagenlc potential of the substance.  Commonly, results from tests
 1n bacterial systems (such as  the Ames test) will be  reported In this section.
 as well  as various  In vitro  tests In eukaryotlc cells.   Scientific guidelines
 to be followed  are  given in  FR Si (18S):3400S.]

 CarclnoqenlcUy

      [This section  should  contain the results of experiments  designed to
 assess the potential,of selected chemicals  of  Interest  to  produce tumors  in
 the various organs/systems.  Scientific  guidelines  to be followed are given  1n
 FR 51 (185);33I92 U.S. EM.  198$.]

 Other Effects

      [If any other  p*rt1ntnt Information is detected  In the literature, it
 should be presented htrt.  For example,  syntrglstle  or  antagonistic  effects
with other compounds May bt  summarized.]
                                       -16-

-------
 VI
  iUf* Of  iQXiCULQGICAL EFFECTS
      Health  Advisories are based upon the Identification of  adverse health
 effects  associated with  the most sensitive nonearclnogenlc endpolnt of
 toxlclty,  The/ Induction of thH/thesi effeetU) 1s/are related  to a
 particular exposure  dose regime over a specified period of  time.  The effect
 1s  usually determined from an animal lexicological study. Standardized risk
 character ization methodology for threshold toxicants Is applied  In HA
 development.   The general formula Is is follows:
      HA
             JNOAEl)  (8W)
      where;  NQAfl



             I OAR


             BM


             UF(S)
             _i/day


1 -day Health Advisory
                   jng/1  (_»g/i)
            No-Observed-Advtrse-iffect  Level
            or
            lowtst-Observed-Adverse Effect Level
            (the exposure dose In mg/kg bw/day}

            assumed body weight of protected Individual
            OO-fcg fof cnlld or 70 kg for an atfult)

            uncertainty factor, chosen using U.S.  IPA,  1986a
            guidelines to compensate the uncertainty of  NOAU
            extrapolations from animal/human studies to human
            populations taking Into consideration  the amount,
            type, and nature of the data

            assumed dally water consumption (1 i/chUd:   2
            t/aduit)
     Tne study by 	
the 10-kg child 1-day MA
Factors to be consldertd
period, age of test animals
measured, experimental
Interactions
           	has been selected to serve as the basis for
           because-..{add reasons why this study was selected.
           Include appropriateness of exposure route, exposure
              (or humans), sensitivity of the parameter
         limitations, etc.  High risk populations, chemical
beneficial effects and other special considerations should also
be Included, as
calculation.}
  appropriate. In the selection of studies and/or the
                                       -17-

-------
 The  i-day HA  for  ins 10 kg chHd U calculate as  follows:

                          (KOAEt)  (10 kg)
      l-day HA
     where:
             NOAEL
             UF
                           (l I/day) (UP)
=  based on absence  of  [effect]  In  [species] exposed to
   [substance]  via  [route]  for [duration]
             10 kg     »  assumed, weight of child

             1 i/day '  *  assumed water consumption'by a 10 kg  child.
   uncertainty factor;  chosen  m  accordance with U.S.
   EPA,  19861 guidelines
(a)In this case and in all other calculations below, if a lOAEl Is  employed,
the test should read; *	w
-------
 Longer-Term Health Advisory

      The study &y
         Has been  selected  to serve as the oasH  for
 the longer-tefin HA because,...[add reasons why this study was  selected]

 The longer-terra HA for  the  10-kg child is calculated as follows:

         LOfger-term HA   •   (KQAEL  i  (10 kg)  *
         when:,

              NQAfl



              10  kf

              UF


              1 I/day
                             {UF)   {1 t/day)
a  (in ug/kg/   assumed dally water consumption by an adult

lifetime Health Advisory

The lifetime HA represents  that portion of an Individual's total exposure that
is attributed to drinking water and  1s considered protective  of noncarclno-
genlc adverse health effects over a  lifetime eiposure.  The lifetime MA  1s
derived In a three step process.  Step 1 determines  the Reference  Dose  (RfD),
formerly called the Acceptable Dally Intake (ADI).   The RfD Is an  estimate
(with uncertainty spanning  perhaps an order of magnitude)  of  a dally exposure
to the human population (including sensitive subgroups) that  1s  likely  to be
without appreciable risk of deleterious health effects during a  lifetime, and
Is derived from the HOAH (or IQAIL). Identified  from a chronic
                                       -19-

-------
 (or  subehronlc)  study, divided by an uncertainty faetor{s) times a modifying
 factor.   From  the. SfO, a Drinking later Equivalent Level (OWil) can be
 determined  (S.tep_*2),  A OWEL  is a medium-specific (i.e., drinking water)
 lifetime  exposure  level, assuming 100% exposure from that nwdlum, at which
 adverse,  nonearclnogenlc health effects would not be expected to occur.  The
 QWEL  is der1j|fd  from  the multiplication of the RfD by the assumed body weight
 of an  adult Hd  divided by  tht assumed dally water consumption of an adult,
 Tht  Ilfet1ro*?ttt  In drinking water ilone is determined In Step 3 by factoring
 in other  source's >f exposure, the relative source contribution (ISC).  The RSC
 from drinking  water is based  on actual exposure data or, if data are not
 available, a value of 20% is  assumed for synthetic organic chemicals and a
 value  of  10% Is  assumed -for Inorganic chemicals.  If me contaminant 1s
 classified as  a  Group A or  8  carcinogen, according to the Agency's
 classification scheme of carcinogenic potential (U.S. EPA. Ii86b)  then caution
 should be exercised in assessing the risks associated with lifetime exposure
 to this chemical,

     The  study by _ . _ has  been selected to serve as  tht basis  for the
 lifetime  HA because ...(state reasons as to why study was selected}.

Using  this study the  lifetime HA Is derived as follows:

Step 1: Determination of the  Reference Dose  (RfD)

        RfD      »       NOAEL (of  LOMU mg/kq/dav » _ rag/kg/day
                            UF(s) x  Hf

        where;

        NQAEl    --       (or LOAEL)  U based  on the  absence of [effect] in
                        [species] exposed  to  [substance]  via [route]  for
                        [duration].

        UF       -       10, 100, 1000, or  10,000  according  to U.S. EPA, 1986a
                        guidelines
        «f      *       Modifying  factor  >0 to <1

Step 2: Determination of  tht  Drinking W*.ter Level  (DHfl)

        OWEL    .       (JtfD)   x   70 kg   . _ nig/i  (
                           (2  I/day)

     where:

        RfD              »   is given  pf/kg/day

        70 kg            >   body weight of protected individual

        2 l/day          =   Assumed dally water consumption by an adult
                                       -20-

-------
 Step 3: Determination of  lifetime  HA  for drinking water only

      Lifetime HA        •   OHR  X  RSC - _ mg/t ( _ wg/i)

      where;  .'

         RSC             =   Relative Source Contribution (10% or 20% unless
                            more  Information Is available that allows more
                            concise determinations)

 If the available data provide  information on the exposure potential from Hsh
 along wHh the*bioconcentrat1on  factors involved, the author win also
 calculate  a health  advisory for  ambient water.  This approach allows one to
 calculate  the lifetime acceptable  concentration In ambient water based on
 intake resulting from the  consumption of nsh only (consumed at i rate of
 0,0065 kg/day) and  that contributed by the consumption of fish plus irtgestlon
 of potable water (2 i/day).

 The two procedures  are described as follows;

      a,  fish  Only

         HA *         (RfD (BW)  _
                 (BCF)(fish consumption)

      b.  Fish  plus potable  water

         HA «         (RfQ (SH) _
                 2 I/day *  (BCF)  (fish Consumption)
        BM                *   body weight  (70 Kg/adult)

        BCF               *   Hoconcentratlon factor  (to be supplied)

     2 I/day              =   water consumed per adult/day

     flsn consumption  *  0.0065 kg/day


Depending upon the amount of  data available, a relative source contribution
may also be utilized lo further refine the HA calculated abovt.  This  is done
my multiplying the final number by the percentage assumed  to  be  contributed  by
ambient water.  The assumptions used  In  these calculations must  be  thoroughly
documented.

Cancer B 1 sjt Asset sgent

[The HA document author Is not required  to develop  the  actual quantitative
carcinogenic risk assessment.  However the author will  report to the EPA any
data that can be used in the  risk assessment calculations  by  the EPA.   from
these data, the IPA can determine the concentration of  the chenlcal in
                                       -21-

-------
 water (mg/i)  that  U  equivalent  to  the exposure (mg/kg/day)  for  each  excess
 cancer risk  level  (10*.  10"*,  10"*).  The basic assumption used  as  well
 as  the general  procedures  followed  In the determination of these risk levels
 should accompany  the  final rtsylts.  If there are no carcinogenic data
 available, a  negative statement  to  that effect should be made.  If the chemical
 has  been classified by |PA or  IARC  (EPA 1s preferred) then the boiler plate
 caveats  stating the chemical's classification should b* included.  The IPA
 reference to  use  is U.S. EPA,  19S6b.


 VII,  OTHEK,CRITERIA,  GUIDANCE  AND STANDARDS

 [IMefly summarize any existing  guidelines by EPA, National Institute for
 Occupational  Safety aed  Health (NlOSH)/Qccupat1onal -Safety and Health
 Administration  (QSHA), Food and  Drug Administration (FQA), other Federal
 agencies,  states and/or  foreign  nations. Including organoleptlc data.  Also
 summarize  any guidelines promulgated by the National Academy of Sciences
 (MAS), world  Health Organization (WHO), etc,]


 VI11.  EPA  CONTACTS

      A.  AQUATIC LIFE  ADVISORIES

         For further Information  regarding the aquatic  life and  fHh  and water
         exposure advisories contact:

            ..__., 		 FTS 47S-7315   (202)475-7315

         	.__.__.	 FTS 475-7315   (202)475-7315
     I. DRINKING WATER ADVISORIES

        For further  Information  regarding  the  drinking  water  human  health
        advisories contact;

                                                   (202J382-	

                                                   (202)382-	
IX, REFERENCES

(This section contains a  listing  of  all  references cited In the HA docyment In
the proper format.

NAS (National Academy of  Sciences).   1977.   Drinking Water and Health. Vol. l,
p. 19-63.

NAS (National Acadtffly of  Sciences).   liBO.   Drinking Water and Health, vol. 3,
p. 25-67,
                                       -22-

-------
u,S. EPA.  I986a,  Appendix A.   Reference Dose (RfD):  Description  ana  use  \n
HeaHn Risk Assessment.  Integrated Risk Information System (IRIS).  Online.
Office of Health and Environmental Assessment, Environmental  Criteria
Assessment Office. Cincinnati,  OH,

U.S. EPA.  1*86b.  Guidelines for Carcinogen Risk Assessment.   Federal
Register 51(18$);3399J-34QQ3.
                                       -23-

-------